Class representing a mixture of PDFs, the mixture can contain different. More...

#include <mixtureParticleFilter.h>

Inheritance diagram for BFL::BFL::Mixture< T >:

Public Member Functions
bool	AddComponent (Pdf< T > &pdf)
	Add a component pdf: THIS IS A NON-REALTIME OPERATION. More...

bool	AddComponent (Pdf< T > &pdf, Probability w)
	Add a component pdf with weight w: THIS IS A NON-REALTIME OPERATION. More...

virtual Mixture *	Clone () const
	Clone function. More...

Pdf< T > *	ComponentGet (unsigned int componentNumber) const
	Get the pointer to the component pdf of component "componentNumber". More...

vector< Pdf< T > * >	ComponentsGet () const
	Get the vector of pointers to the component pdfs. More...

MatrixWrapper::SymmetricMatrix	CovarianceGet () const
	Get the Covariance Matrix E[(x - E[x])^2] of the Analytic pdf. More...

bool	DeleteComponent (unsigned int componentNumber)
	Delete a component pdf: THIS IS A NON_REALTIME OPERATION. More...

template<>
ColumnVector	ExpectedValueGet () const
	Get the expected value E[x] of the pdf. More...

template<>
unsigned int	ExpectedValueGet () const
	Get the expected value E[x] of the pdf. More...

template<>
int	ExpectedValueGet () const
	Get the expected value E[x] of the pdf. More...

T	ExpectedValueGet () const
	Get the expected value E[x] of the pdf. More...

template<>
double	ExpectedValueGet () const
	Get the expected value E[x] of the pdf. More...

	Mixture (const unsigned int dimension=0)
	Constructor: An equal weight is set for all components. More...

template<typename U >
	Mixture (const U &componentVector)
	Constructor: An equal weights is set for all components. More...

	Mixture (const Mixture &my_mixture)
	Copy Constructor. More...

int	MostProbableComponentGet () const
	Get the index of the most probable component, if a few component are. More...

unsigned int	NumComponentsGet () const
	Get the number of components. More...

Probability	ProbabilityGet (const T &state) const
	Implementation of virtual base class method. More...

bool	SampleFrom (vector< Sample< T > > &list_samples, const unsigned int num_samples, int method=DEFAULT, void *args=NULL) const
	Draw multiple samples from the Pdf (overloaded) More...

bool	SampleFrom (Sample< T > &one_sample, int method=DEFAULT, void *args=NULL) const
	Draw 1 sample from the Pdf: More...

Probability	WeightGet (unsigned int componentNumber) const
	Get the component weight of component "componentNumber". More...

bool	WeightSet (unsigned int componentNumber, Probability w)
	Function to change/set the weigth of a single component. More...

vector< Probability >	WeightsGet () const
	Get all component weights. More...

bool	WeightsSet (vector< Probability > &weights)
	Set all component weights. More...

virtual	~Mixture ()
	Destructor. More...

Public Member Functions inherited from BFL::BFL::Pdf< T >
unsigned int	DimensionGet () const
	Get the dimension of the argument. More...

virtual void	DimensionSet (unsigned int dim)
	Set the dimension of the argument. More...

	Pdf (unsigned int dimension=0)
	Constructor. More...

virtual	~Pdf ()
	Destructor. More...

Protected Member Functions
bool	CumWeightsUpdate ()
	Updates the cumWeights. More...

bool	NormalizeWeights ()
	Normalize the component weigths (eg. after setting a component weight) More...

void	TestNotInit () const
	Called when a the number of components=0 and if method is called which. More...

Protected Attributes
vector< Pdf< T > * > *	_componentPdfs
	Pointer to the vector of component pdfs. More...

vector< Probability > *	_componentWeights
	Pointer to the vector of mixture weights, the sum of the elements = 1. More...

vector< double >	_cumWeights
	Vector containing the cumulative component weights (for efficient sampling) More...

unsigned int	_numComponents
	The number of components. More...

Detailed Description

template<typename T>
class BFL::BFL::Mixture< T >

Class representing a mixture of PDFs, the mixture can contain different.

This class is an instantation from the template class Pdf, with added methods to get a set the individual components and mixture weights

Definition at line 48 of file mixtureParticleFilter.h.

Constructor & Destructor Documentation

template<typename T >

BFL::Mixture< T >::Mixture ( const unsigned int dimension = 0 )

Constructor: An equal weight is set for all components.

Parameters

dimension the dimension of the state space

Definition at line 182 of file mixtureParticleFilter.h.

template<typename T >

template<typename U >

BFL::Mixture< T >::Mixture ( const U & componentVector )

Constructor: An equal weights is set for all components.

Parameters

componentVector vector of components (Pdf<T>*)

Definition at line 199 of file mixtureParticleFilter.h.

template<typename T >

BFL::Mixture< T >::Mixture ( const Mixture< T > & my_mixture )

Copy Constructor.

Definition at line 226 of file mixtureParticleFilter.h.

template<typename T >

BFL::Mixture< T >::~Mixture ( )

virtual

Destructor.

Definition at line 247 of file mixtureParticleFilter.h.

Member Function Documentation

template<typename T >

bool BFL::Mixture< T >::AddComponent ( Pdf< T > & pdf )

Add a component pdf: THIS IS A NON-REALTIME OPERATION.

Parameters

pdf	Component pdf which will be added

Definition at line 467 of file mixtureParticleFilter.h.

template<typename T >

bool BFL::Mixture< T >::AddComponent	(	Pdf< T > &	pdf,
		Probability	w
	)

Add a component pdf with weight w: THIS IS A NON-REALTIME OPERATION.

Parameters

pdf	Component pdf which will be added
weight	the weight of the new component

Definition at line 487 of file mixtureParticleFilter.h.

template<typename T >

Mixture< T > * BFL::Mixture< T >::Clone ( ) const

virtual

Clone function.

Implements BFL::BFL::Pdf< T >.

Definition at line 262 of file mixtureParticleFilter.h.

template<typename T >

Pdf< T > * BFL::Mixture< T >::ComponentGet ( unsigned int componentNumber ) const

Get the pointer to the component pdf of component "componentNumber".

Parameters

componentNumber number of the component (must be >=0 and <_numComponents)

Definition at line 540 of file mixtureParticleFilter.h.

template<typename T >

vector< Pdf< T > * > BFL::Mixture< T >::ComponentsGet ( ) const

Get the vector of pointers to the component pdfs.

Definition at line 533 of file mixtureParticleFilter.h.

template<typename T >

MatrixWrapper::SymmetricMatrix BFL::Mixture< T >::CovarianceGet ( ) const

virtual

Get the Covariance Matrix E[(x - E[x])^2] of the Analytic pdf.

Get first order statistic (Covariance) of this AnalyticPdf

Returns: The Covariance of the Pdf (a SymmetricMatrix of dim DIMENSION)

Todo:: extend this more general to n-th order statistic

Bug:: Discrete pdfs should not be able to use this!

Reimplemented from BFL::BFL::Pdf< T >.

Definition at line 378 of file mixtureParticleFilter.h.

template<typename T >

bool BFL::Mixture< T >::CumWeightsUpdate ( )

protected

Updates the cumWeights.

Definition at line 577 of file mixtureParticleFilter.h.

template<typename T >

bool BFL::Mixture< T >::DeleteComponent ( unsigned int componentNumber )

Delete a component pdf: THIS IS A NON_REALTIME OPERATION.

Parameters

componentNumber number of the component which will be deleted (must be >=0 and <_numComponents)

Definition at line 507 of file mixtureParticleFilter.h.

template<>

ColumnVector BFL::BFL::Mixture< ColumnVector >::ExpectedValueGet ( ) const

inlinevirtual

Get the expected value E[x] of the pdf.

Get low order statistic (Expected Value) of this AnalyticPdf

Returns: The Expected Value of the Pdf (a ColumnVector with DIMENSION rows)

Note: No set functions here! This can be useful for analytic functions, but not for sample based representations!; For certain discrete Pdfs, this function has no meaning, what is the average between yes and no?

Reimplemented from BFL::BFL::Pdf< T >.

Definition at line 31 of file mixtureParticleFilter.h.

template<>

unsigned int BFL::BFL::Mixture< unsigned int >::ExpectedValueGet ( ) const

inlinevirtual

Get the expected value E[x] of the pdf.

Get low order statistic (Expected Value) of this AnalyticPdf

Returns: The Expected Value of the Pdf (a ColumnVector with DIMENSION rows)

Note: No set functions here! This can be useful for analytic functions, but not for sample based representations!; For certain discrete Pdfs, this function has no meaning, what is the average between yes and no?

Reimplemented from BFL::BFL::Pdf< T >.

Definition at line 46 of file mixtureParticleFilter.h.

template<>

int BFL::BFL::Mixture< int >::ExpectedValueGet ( ) const

inlinevirtual

Get the expected value E[x] of the pdf.

Get low order statistic (Expected Value) of this AnalyticPdf

Returns: The Expected Value of the Pdf (a ColumnVector with DIMENSION rows)

Note: No set functions here! This can be useful for analytic functions, but not for sample based representations!; For certain discrete Pdfs, this function has no meaning, what is the average between yes and no?

Reimplemented from BFL::BFL::Pdf< T >.

Definition at line 82 of file mixtureParticleFilter.h.

template<typename T >

T BFL::Mixture< T >::ExpectedValueGet ( ) const

virtual

Get the expected value E[x] of the pdf.

Get low order statistic (Expected Value) of this AnalyticPdf

Returns: The Expected Value of the Pdf (a ColumnVector with DIMENSION rows)

Note: No set functions here! This can be useful for analytic functions, but not for sample based representations!; For certain discrete Pdfs, this function has no meaning, what is the average between yes and no?

Reimplemented from BFL::BFL::Pdf< T >.

Definition at line 368 of file mixtureParticleFilter.h.

template<>

double BFL::BFL::Mixture< double >::ExpectedValueGet ( ) const

inlinevirtual

Get the expected value E[x] of the pdf.

Get low order statistic (Expected Value) of this AnalyticPdf

Returns: The Expected Value of the Pdf (a ColumnVector with DIMENSION rows)

Note: No set functions here! This can be useful for analytic functions, but not for sample based representations!; For certain discrete Pdfs, this function has no meaning, what is the average between yes and no?

Reimplemented from BFL::BFL::Pdf< T >.

Definition at line 116 of file mixtureParticleFilter.h.

template<typename T >

int BFL::Mixture< T >::MostProbableComponentGet ( ) const

Get the index of the most probable component, if a few component are.

Definition at line 450 of file mixtureParticleFilter.h.

template<typename T >

bool BFL::Mixture< T >::NormalizeWeights ( )

protected

Normalize the component weigths (eg. after setting a component weight)

Definition at line 558 of file mixtureParticleFilter.h.

template<typename T >

unsigned int BFL::Mixture< T >::NumComponentsGet ( ) const

Get the number of components.

Definition at line 268 of file mixtureParticleFilter.h.

template<typename T >

Probability BFL::Mixture< T >::ProbabilityGet ( const T & state ) const

virtual

Implementation of virtual base class method.

Reimplemented from BFL::BFL::Pdf< T >.

Definition at line 274 of file mixtureParticleFilter.h.

template<typename T >

bool BFL::Mixture< T >::SampleFrom	(	vector< Sample< T > > &	list_samples,
		const unsigned int	num_samples,
		int	method = `DEFAULT`,
		void *	args = `NULL`
	)		const

virtual

Draw multiple samples from the Pdf (overloaded)

Parameters

list_samples	list of samples that will contain result of sampling
num_samples	Number of Samples to be drawn (iid)
method	Sampling method to be used. Each sampling method is currently represented by a #define statement, eg. #define BOXMULLER 1
args	Pointer to a struct representing extra sample arguments. "Sample Arguments" can be anything (the number of steps a gibbs-iterator should take, the interval width in MCMC, ... (or nothing), so it is hard to give a meaning to what exactly Sample Arguments should represent...

Todo:: replace the C-call "void * args" by a more object-oriented structure: Perhaps something like virtual Sample * Sample (const int num_samples,class Sampler)

Bug:: Sometimes the compiler doesn't know which method to choose!

Reimplemented from BFL::BFL::Pdf< T >.

Definition at line 286 of file mixtureParticleFilter.h.

template<typename T >

bool BFL::Mixture< T >::SampleFrom	(	Sample< T > &	one_sample,
		int	method = `DEFAULT`,
		void *	args = `NULL`
	)		const

virtual

Draw 1 sample from the Pdf:

There's no need to create a list for only 1 sample!

Parameters

one_sample	sample that will contain result of sampling
method	Sampling method to be used. Each sampling method is currently represented by a #define statement, eg. #define BOXMULLER 1
args	Pointer to a struct representing extra sample arguments

See also: SampleFrom()

Bug:: Sometimes the compiler doesn't know which method to choose!

Reimplemented from BFL::BFL::Pdf< T >.

Definition at line 339 of file mixtureParticleFilter.h.

template<typename T >

void BFL::Mixture< T >::TestNotInit ( ) const

protected

Called when a the number of components=0 and if method is called which.

Definition at line 547 of file mixtureParticleFilter.h.

template<typename T >

Probability BFL::Mixture< T >::WeightGet ( unsigned int componentNumber ) const

Get the component weight of component "componentNumber".

Parameters

componentNumber number of the component (must be >=0 and <_numComponents)

Definition at line 397 of file mixtureParticleFilter.h.

template<typename T >

bool BFL::Mixture< T >::WeightSet	(	unsigned int	componentNumber,
		Probability	w
	)

Function to change/set the weigth of a single component.

Changes the component weights such that AFTER normalization the weight of the component "componentNumber" is equal to the weight w

Parameters

componentNumber	number of the component of which the weight will be set (must be >=0 and <_numComponents)
w	Probability to which the weight of component "componentNumber" will be set (must be <= 1)

Definition at line 415 of file mixtureParticleFilter.h.

template<typename T >

vector< Probability > BFL::Mixture< T >::WeightsGet ( ) const

Get all component weights.

Definition at line 390 of file mixtureParticleFilter.h.

template<typename T >

bool BFL::Mixture< T >::WeightsSet ( vector< Probability > & weights )

Set all component weights.

values vector<Probability> containing the new component weights . The sum of the probabilities of this list is not required to be one since the normalization is automatically carried out. The size of weights should be equal to the number of components

Definition at line 405 of file mixtureParticleFilter.h.

Member Data Documentation

template<typename T>

vector<Pdf<T>* >* BFL::BFL::Mixture< T >::_componentPdfs

protected

Pointer to the vector of component pdfs.

Definition at line 58 of file mixtureParticleFilter.h.

template<typename T>

vector<Probability>* BFL::BFL::Mixture< T >::_componentWeights

protected

Pointer to the vector of mixture weights, the sum of the elements = 1.

Definition at line 55 of file mixtureParticleFilter.h.

template<typename T>

vector<double> BFL::BFL::Mixture< T >::_cumWeights

protected

Vector containing the cumulative component weights (for efficient sampling)

Definition at line 66 of file mixtureParticleFilter.h.

template<typename T>

unsigned int BFL::BFL::Mixture< T >::_numComponents

protected

The number of components.

Definition at line 52 of file mixtureParticleFilter.h.

The documentation for this class was generated from the following files:

Public Member Functions

Protected Member Functions

Protected Attributes

Detailed Description

template<typename T> class BFL::BFL::Mixture< T >

Constructor & Destructor Documentation

Member Function Documentation

Member Data Documentation

template<typename T>
class BFL::BFL::Mixture< T >